ggml update to b7108 (#12992)

* Revert "vulkan: temporary cary of vulkan fixes (#12971)" This reverts commit 3a9e8e9f. * ggml update to b7087 * fix argsort on metal * update to b7108 * fix bakllava regression This model lacks the metadata for the projector type. * update to b7209 * fix TopK perf * only build arm code on arm

ggml update to b7108 (#12992)
* Revert "vulkan: temporary cary of vulkan fixes (#12971)" This reverts commit 3a9e8e9f. * ggml update to b7087 * fix argsort on metal * update to b7108 * fix bakllava regression This model lacks the metadata for the projector type. * update to b7209 * fix TopK perf * only build arm code on arm
0cf7794b · Daniel Hiltgen · GitHub · 854d40ed · 0cf7794b · 0cf7794b
Unverified Commit 0cf7794b authored Dec 03, 2025 by Daniel Hiltgen Committed by GitHub Dec 03, 2025
20 changed files
--- a/llama/llama.cpp/tools/mtmd/clip-impl.h
+++ b/llama/llama.cpp/tools/mtmd/clip-impl.h
--- a/llama/llama.cpp/tools/mtmd/clip.cpp
+++ b/llama/llama.cpp/tools/mtmd/clip.cpp
--- a/llama/llama.cpp/tools/mtmd/clip.h
+++ b/llama/llama.cpp/tools/mtmd/clip.h
--- a/llama/llama.cpp/tools/mtmd/mtmd-helper.cpp
+++ b/llama/llama.cpp/tools/mtmd/mtmd-helper.cpp
--- a/llama/llama.cpp/tools/mtmd/mtmd-helper.h
+++ b/llama/llama.cpp/tools/mtmd/mtmd-helper.h
--- a/llama/llama.cpp/tools/mtmd/mtmd.cpp
+++ b/llama/llama.cpp/tools/mtmd/mtmd.cpp
--- a/llama/llama.cpp/tools/mtmd/mtmd.h
+++ b/llama/llama.cpp/tools/mtmd/mtmd.h
--- a/llama/patches/0001-ggml-backend-malloc-and-free-using-the-same-compiler.patch
+++ b/llama/patches/0001-ggml-backend-malloc-and-free-using-the-same-compiler.patch
--- a/llama/patches/0002-pretokenizer.patch
+++ b/llama/patches/0002-pretokenizer.patch
--- a/llama/patches/0003-clip-unicode.patch
+++ b/llama/patches/0003-clip-unicode.patch
--- a/llama/patches/0004-solar-pro.patch
+++ b/llama/patches/0004-solar-pro.patch
--- a/llama/patches/0005-fix-deepseek-deseret-regex.patch
+++ b/llama/patches/0005-fix-deepseek-deseret-regex.patch
--- a/llama/patches/0006-maintain-ordering-for-rules-for-grammar.patch
+++ b/llama/patches/0006-maintain-ordering-for-rules-for-grammar.patch
--- a/llama/patches/0008-add-phony-target-ggml-cpu-for-all-cpu-variants.patch
+++ b/llama/patches/0008-add-phony-target-ggml-cpu-for-all-cpu-variants.patch
--- a/llama/patches/0009-remove-amx.patch
+++ b/llama/patches/0009-remove-amx.patch
--- a/llama/patches/0010-fix-string-arr-kv-loading.patch
+++ b/llama/patches/0010-fix-string-arr-kv-loading.patch
--- a/llama/patches/0011-ollama-debug-tensor.patch
+++ b/llama/patches/0011-ollama-debug-tensor.patch
--- a/llama/patches/0012-add-ollama-vocab-for-grammar-support.patch
+++ b/llama/patches/0012-add-ollama-vocab-for-grammar-support.patch
--- a/llama/patches/0013-add-argsort-and-cuda-copy-for-i32.patch
+++ b/llama/patches/0013-add-argsort-and-cuda-copy-for-i32.patch
--- a/llama/patches/0014-graph-memory-reporting-on-failure.patch
+++ b/llama/patches/0014-graph-memory-reporting-on-failure.patch